Abstract

A rat central venous catheter (CVC) infection model was used to assess the importance of the proteinacious autolysin (AtlE) and the polysaccharide intercellular adhesin (PIA) in the pathogenesis of Staphylococcus epidermidis CVC-associated infection. Wild-type (wt) S. epidermidis O-47 was significantly more likely to cause a CVC infection than was either of the isogenic mutant strains (AtlE-negative [O-47mut1] or PIA-negative [O-47mut2]). Bacteria were retrieved from the explanted catheters of 87.5% of rats inoculated with S. epidermidis O-47, compared with 25% of rats challenged with either S. epidermidis O-47mut1 or O-47mut2 (P=.007). Peripheral bacteremia was documented in 75% of rats challenged with S. epidermidis O-47, compared with 12.5% and 25% challenged with O-47mut1 and O-47mut2, respectively (P=.009). Metastatic disease was more common in rats inoculated with wt S. epidermidis compared with AtlE- or PIA-deficient mutants. These results confirm the importance of initial adherence, associated with AtlE, and biofilm production, mediated by PIA, in the pathogenesis of S. epidermidis experimental CVC infection

Staphylococcus epidermidis is a commensal inhabitant of human skin that rarely causes disease in healthy persons. In recent years, however, largely because of the increased use of intravascular catheters and other indwelling prosthetic devices, S. epidermidis has emerged as a major nosocomial pathogen [1, 2]. According to the Centers for Disease Control and Prevention National Nosocomial Infection Surveillance System, during the 1990s, coagulase-negative staphylococci accounted for 37.3% of nosocomial bloodstream infections [3]. Unfortunately, these infections are responsible for significant morbidity and mortality [4–6]

The propensity for S. epidermidis to cause intravascular catheter–associated infection can be explained largely by its ability to adhere and proliferate on polymer surfaces. The adherence process for S. epidermidis appears to be a 2-step process that requires adhesion of the bacteria to the biomaterial surface, followed by cell-to-cell accumulation and biofilm formation

In previous studies, Heilmann et al. [7, 8] demonstrated that a proteinacious autolysin (AtlE) is associated with initial attachment of S. epidermidis to plastic. Mack et al. [9–11] described a β-1,6 linked glucosaminoglycan intercellular adhesin (polysaccharide intercellular adhesin [PIA]), synthesized by gene products of the ica locus, that mediates cell-to-cell adherence and biofilm accumulation. We demonstrated that PIA also mediates hemagglutination and is essential in the pathogenesis of foreign body infections in 2 in vivo models [12–14]. In the present study, we evaluated the relative contribution of PIA and AtlE in an isogenic background in the pathogenesis of intravascular catheter–associated infection in a rat model

Materials and Methods

Bacteria.Staphylococcus epidermidis O-47 is a biofilm-positive clinical isolate recovered from a patient at the Institute Für Medizinische Mikrobiologie und Hygiene, Universität zu Köln (Köln, Germany) [7]. S. epidermidis O-47mut1 is a Tn917 insertional mutant that is deficient in AtlE production [8]. S. epidermidis O-47mut2 is a Tn917 insertional mutant that is deficient in PIA production because of interruption of the ica locus [15]

Rat central venous catheter (CVC)–associated infection modelA rat CVC-associated infection model was used to compare the virulence of S. epidermidis O-47 with its isogenic mutants S. epidermidis O-47mut1 and S. epidermidis O-47mut2. Twenty-four male Sprague Dawley rats (Charles River) underwent catheterization, as described elsewhere [14, 16]. In brief, a silastic catheter was inserted into the jugular vein and was advanced into the superior vena cava. The proximal portion of the catheter was tunneled subcutaneously to exit in the midscapular space. A rodent restraint jacket was used to protect the catheter and to allow access to it. Twenty-four hours after CVC placement, blood cultures were obtained to ensure sterility, the rats were divided into groups of 8, and 107 cfu of S. epidermidis O-47 or one of the isogenic mutants was injected into the CVC. The catheters were flushed daily with a heparin solution. On day 8, the rats were killed. Peripheral blood was obtained and was cultured quantitatively by directly plating 0.1 mL of blood on trypticase soy agar (TSA; Remel). The location of the distal tip of the CVC in the superior vena cava was confirmed, the CVC and surrounding venous tissues were removed aseptically and were vigorously vortex washed in PBS, after which the wash fluid was quantitatively cultured. Previous studies, in which quantitative culture results were confirmed by electron microscopy, documented complete removal of adherent organisms by this procedure [13]. In addition, to ascertain the extent of metastatic disease, the heart, lungs, liver, and kidneys were aseptically harvested, were weighed, were homogenized in 1.0 mL of PBS with a mechanical tissue homogenizer (PowerGen 700D; Fisher) and were cultured quantitatively by plating 0.1 mL of tissue homogenate on TSA. Bacteria recovered from the catheter, blood, or tissues were identified to species level

To limit the number of experimental rats used to the minimum that would reveal a significant difference in pathogenicity, inoculum size experiments were performed. Before conducting the comparative studies described above, the smallest inoculum that reliably resulted in a CVC-associated infection was determined in dose-ranging inoculum studies. CVCs were placed as described above. Three rats each were challenged with an inoculum of either 103, 104, 105, 106, or 107 cfu of S. epidermidis O-47. After inoculation, the rats were evaluated for the presence of infection, as described above

To document the importance of the intravascular catheter in the pathogenesis of infection, 4 rats, without indwelling CVCs, were inoculated via the tail vein with the minimal infectious dose of S. epidermidis O-47, as determined above. After inoculation, the rats were evaluated for evidence of infection, as described above

Statistical analysisNonparametric methods were employed to compare the bacterial burden among the groups of experimental rats inoculated with either S. epidermidis O-47 or one of the isogenic mutants, mut 1 or mut 2. The bacterial counts obtained from explanted CVCs, peripheral blood, and peripheral tissues for the 3 groups of rats were compared by using the Kruskal-Wallis test, which was performed using the SAS system (SAS Institute)

Results

Inoculation studiesThe inoculation studies showed that rats challenged with 103 cfu exhibited no sign of CVC-associated infection at the time of death. One of 3 rats challenged with 104 or 105 cfu of S. epidermidis O-47 exhibited evidence of infection at the time of death, compared with 2 of 3 and 3 of 3 rats challenged with 106 and 107 cfu, respectively. Therefore, the 107 cfu inoculum size, the lowest level that reproducibly caused CVC-associated infection and metastatic disease, was used in the larger comparative trial. None of the 4 rats inoculated with 107 cfu of S. epidermidis O-47 via the tail vein had evidence of bacteremia or metastatic disease at the time of death

Comparative studiesSeven (87.5%) of 8 rats inoculated with S. epidermidis O-47 had bacteria recovered from the explanted CVC, compared with 2 of 8 rats challenged with either S. epidermidis O-47mut1 or S. epidermidis O-47mut2. The number of bacteria recovered from the explanted catheters was significantly different among the different groups of rats (P=.007). The number of bacteria recovered from the catheters is shown in figure 1

Figure 1

Recovery of Staphylococcus epidermidis O-47 or its isogenic autolysin-deficient mutant (mut1) or polysaccharide intercellular adhesin (PIA)–deficient mutant (mut2) from explanted central venous catheters (CVCs) in the rat CVC-associated infection model. Median number of organisms (colony-forming units [cfu]/catheter [cath]) recovered from the CVC of rats inoculated with wild-type S. epidermidis O-47 was 6.1×102 cfu (range, 0–7.8×103 cfu; lower/upper quartiles, 9–1.1×103 cfu), compared with a median of 0 cfu (range, 0–10 cfu; lower/upper quartiles, 0–5 cfu) for the autolysin-deficient mutant and a median of 0 cfu (range, 0–220 cfu; lower/upper quartiles, 0–5 cfu) for the PIA-deficient mutant (P=.007, Kruskal-Wallis test; P=.027, corrected P value for S. epidermidis O-47 vs. S. epidermidis O-47mut1; P=.045, S. epidermidis O-47 vs. S. epidermidis O-47mut2). There was no significant difference between the mut1 and mut2 mutant strains (P=.50). Each data point represents cfu/cath for a single experimental animal. Horizontal line median value

Figure 1

Recovery of Staphylococcus epidermidis O-47 or its isogenic autolysin-deficient mutant (mut1) or polysaccharide intercellular adhesin (PIA)–deficient mutant (mut2) from explanted central venous catheters (CVCs) in the rat CVC-associated infection model. Median number of organisms (colony-forming units [cfu]/catheter [cath]) recovered from the CVC of rats inoculated with wild-type S. epidermidis O-47 was 6.1×102 cfu (range, 0–7.8×103 cfu; lower/upper quartiles, 9–1.1×103 cfu), compared with a median of 0 cfu (range, 0–10 cfu; lower/upper quartiles, 0–5 cfu) for the autolysin-deficient mutant and a median of 0 cfu (range, 0–220 cfu; lower/upper quartiles, 0–5 cfu) for the PIA-deficient mutant (P=.007, Kruskal-Wallis test; P=.027, corrected P value for S. epidermidis O-47 vs. S. epidermidis O-47mut1; P=.045, S. epidermidis O-47 vs. S. epidermidis O-47mut2). There was no significant difference between the mut1 and mut2 mutant strains (P=.50). Each data point represents cfu/cath for a single experimental animal. Horizontal line median value

Six (75%) of 8 rats challenged with the wild-type (wt) strain S. epidermidis O-47 had bacteria recovered from the peripheral blood at the time of death, compared with 1 of 8 rats challenged with S. epidermidis O-47mut1 and 2 of 8 rats challenged with S. epidermidis O-47mut2. The number of bacteria recovered from the peripheral blood was significantly different among the different groups of rats (P=.009). In no case were peripheral blood cultures positive without concordant recovery of organisms from the CVC. In 2 rats, the CVC culture was positive, and the peripheral blood culture was negative. In both instances, the number of bacteria recovered from the CVC was ⩽10 cfu. The number of bacteria recovered in the peripheral blood is shown in figure 2

Figure 2

Recovery of Staphylococcus epidermidis O-47 or its isogenic autolysin-deficient mutant (mut1) or polysaccharide intercellular adhesin (PIA)–deficient mutant (mut2) from the peripheral bloodstream in the rat central venous catheter (CVC)–associated infection model. Median number of organisms (colony-forming units [cfu]/mL) recovered from the bloodstream of rats inoculated with wild-type S. epidermidis O-47 was 90 cfu (range, 0–9.2×104 cfu; lower/upper quartiles, 35–1.7×103 cfu), compared with a median of 0 cfu (range, 0–2.6×102 cfu; lower/upper quartiles, 0–0 cfu) for the autolysin-deficient mutant and a median of 0 cfu (range, 0–70 cfu; lower/upper quartiles, 0–20 cfu) for the PIA-deficient mutant (P=.009, Kruskal-Wallis test; P=.053, corrected P value for S. epidermidis O-47 vs. S. epidermidis O-47mut1; P=.057, S. epidermidis O-47 vs. S. epidermidis O-47mut2). There was no significant difference between the mut1 and mut2 mutant strains (P=1.0). Each data point represents cfu/mL blood for a single experimental animal. Horizontal line median value

Figure 2

Recovery of Staphylococcus epidermidis O-47 or its isogenic autolysin-deficient mutant (mut1) or polysaccharide intercellular adhesin (PIA)–deficient mutant (mut2) from the peripheral bloodstream in the rat central venous catheter (CVC)–associated infection model. Median number of organisms (colony-forming units [cfu]/mL) recovered from the bloodstream of rats inoculated with wild-type S. epidermidis O-47 was 90 cfu (range, 0–9.2×104 cfu; lower/upper quartiles, 35–1.7×103 cfu), compared with a median of 0 cfu (range, 0–2.6×102 cfu; lower/upper quartiles, 0–0 cfu) for the autolysin-deficient mutant and a median of 0 cfu (range, 0–70 cfu; lower/upper quartiles, 0–20 cfu) for the PIA-deficient mutant (P=.009, Kruskal-Wallis test; P=.053, corrected P value for S. epidermidis O-47 vs. S. epidermidis O-47mut1; P=.057, S. epidermidis O-47 vs. S. epidermidis O-47mut2). There was no significant difference between the mut1 and mut2 mutant strains (P=1.0). Each data point represents cfu/mL blood for a single experimental animal. Horizontal line median value

Table 1 summarizes results from studies defining the burden of metastatic disease in the rats challenged with the various strains of S. epidermidis. For all organ systems examined, there were more rats with metastatic disease in the group inoculated with wt S. epidermidis O-47 than in the groups inoculated with either adhesin-deficient strain. In addition, for all organ systems there was a greater microbial burden of metastatic disease (as documented by quantitative cultures) in the rats inoculated with the wt strain than in the rats inoculated with the autolysin- or PIA-deficient mutant strains. In general, the rats inoculated with the wt parent strain had 10–100-fold greater numbers of organisms recovered from peripheral organs than did rats inoculated with the adhesin-deficient mutants. There were no significant differences between the groups of rats inoculated with the S. epidermidis O-47mut1 or S. epidermidis O-47mut2 mutant strains

Table 1

Metastatic disease caused by Staphylococcus epidermidis O-47, S. epidermidis O-47mut1, or S. epidermidis O-47mut2 in the rat central venous catheter–associated infection model

Table 1

Metastatic disease caused by Staphylococcus epidermidis O-47, S. epidermidis O-47mut1, or S. epidermidis O-47mut2 in the rat central venous catheter–associated infection model

Discussion

Coagulase-negative staphylococci, predominantly Staphylococcus epidermidis are the most common cause of infections of prosthetic medical devices [2, 17]. It has long been thought that the ability of S. epidermidis to attach to biomaterials and to elaborate biofilm explains its propensity to cause biomaterial-based infections. Bacterial adherence to biomaterials is a complex multistep process that Gristina [18] conveniently subdivided into the stages of attachment, adhesion, and aggregation. Electron microscopic studies have shown that bacteria first attach to the surface and later form colonies and multilayered cell clusters embedded in a thick slime matrix known as biofilm [19]

Attachment is thought to be due to nonspecific factors, such as hydrophobicity and surface charge. Adhesion appears to be secondary to more specific factors. Several investigators have described specific molecular entities that are involved in adhesion. Heilmann et al. [7] described Tn917 insertional mutants of S. epidermidis deficient in their ability to adhere to polystyrene. The phenotype of the mutant was restored in complementation experiments that revealed a 60-kDa protein as essential for adhesion [7]. Further genetic analysis revealed that the protein is encoded by the gene atlE, an autolysin [8]. In addition to mediating adhesion to plastic, AtlE also binds to vitronectin [8]. Polysaccharide adhesin (PSA), described by Pier et al. [20], appears to be important in adherence to plastics. Antibody directed against PSA is protective in animal models of infection, and mutants deficient in PSA are less virulent [21, 22]. McKenney et al. [23] reported data indicating that the ica locus plays a role in the production of PSA and that PSA mediates both initial attachment and aggregation. Timmerman et al. [24, 25] described a 220-kDa cell wall–associated protein (Ssp1) that appears to be functional in adhesion and is organized in a fimbria-like structure

The aggregative phase of adherence, characterized by cell-to-cell adherence and the elaboration of biofilm, is mediated by PIA, as described initially by Mack et al. [11]. The expression of PIA is dependent on glucose in the growth media [11]. Isogenic PIA-negative, Tn917 mutants are capable of initial adhesion but are unable to form microcolonies [10]. Hussain et al. [26] described a 140-kDa protein, termed accumulation-associated protein, that appears to be important in the aggregative stage of adherence

Bacterial adherence to biomaterials, however, is only one component of a complex process that involves interactions between the host, the biomaterial, and the microbe. Therefore, it is desirable to test questions regarding pathogenesis in models that reflect its complexity. We examined the relative contribution of the autolysin, AtlE, and PIA in an isogenic background strain S. epidermidis O-47, in the pathogenesis of intravascular device infections in the rat model of CVC-associated infection. In this study, we added additional support for the multistep model of bacterial adherence and demonstrate the crucial role of adherence in biomaterial-based infection. We observed that a strain of S. epidermidis lacking either the ability to adhere initially to a catheter or the ability to form cell clusters is less virulent than the wt parent strain. It is interesting that we also found that the background strain S. epidermidis O-47 produces less biofilm, as evidenced by hemagglutination titer and biofilm assay (data not shown), than did other S. epidermidis strains we have previously studied (S. epidermidis SE5 and S. epidermidis 1457) and was less virulent in the rat model of CVC-associated infection, as evidenced by the need for a higher infective dose for S. epidermidis O-47 (107 cfu) than S. epidermidis 1457 (104 cfu) [14]. Therefore, not only the presence of adhesins but their relative amount may explain differences in pathogenicity. Further studies should be directed at discerning the interaction, relative importance, and regulation of the adhesins of S. epidermidis

We documented that S. epidermidis O-47 was more likely to cause a CVC-associated infection, manifested by bacteremia and organisms recovered from the explanted catheter (primary end points in CVC-associated infection model), than were the adhesin-deficient mutants. Although the burden of metastatic disease (secondary end points in the CVC-associated infection model) tended to be greater in rats infected with the wt strain, in general, these differences were not statistically significant; however, observations regarding metastatic seeding are limited by the relatively small sample size. Metastatic seeding of organs may be governed by undefined, nonspecific, or tissue-specific adhesins. Perdreau-Remington et al. [27] noted that a chemically induced mutant deficient in biofilm formation showed no decreased virulence in a rabbit model of endocarditis

In conclusion, in an isogenic background strain of S. epidermidis we demonstrated that adhesion and aggregation are essentially equally important in the pathogenesis of intravascular catheter–associated infection in the rat model. Strains lacking in the autolysin, AtlE, or PIA were less likely to cause CVC-associated infection manifest by bacteria recovered from catheter tips, blood, or peripheral organs. These findings may suggest novel means to prevent or treat prosthetic device infections

Acknowledgments

We thank Joseph Ulphani (Dept. of Internal Medicine, University of Nebraska Medical Center, Omaha) for expert technical assistance and James Anderson (Dept. of Preventive and Societal Medicine, University of Nebraska Medical Center) for assistance with the statistical analysis of data

References

1
Boyce
JM
Mayhall
C
Coagulase-negative staphylococci
Hospital epidemiology and infection control
 , 
1999
2d ed
Philadelphia
Lippincott Williams and Wilkins
(pg. 
365
-
83
)
2
Rupp
ME
Archer
GL
Coagulase-negative staphylococci: pathogens associated with medical progress
Clin Infect Dis
 , vol. 
19
 (pg. 
231
-
45
)
3
Centers for Disease Control and Prevention
National nosocomial infections surveillance (NNIS) system report: data summary from January 1990–May 1999, issued June 1999
Am J Infect Control
 , vol. 
27
 (pg. 
520
-
32
)
4
Martin
MA
Pfaller
MA
Wenzel
RP
Coagulase-negative staphylococcal bacteremia: mortality and hospital stay
Ann Intern Med
 , vol. 
110
 (pg. 
9
-
16
)
5
Pittet
D
Tarara
D
Wenzel
RP
Nosocomial bloodstream infection in critically ill patients: excess length of stay, extra costs, and attributable mortality
JAMA
 , vol. 
271
 (pg. 
1598
-
1601
)
6
Rupp
ME
Crossley
KB
Archer
GL
Infections of intravascular catheters and vascular devices
The staphylococci in human disease
 (pg. 
379
-
99
)
7
Heilmann
C
Gerke
C
Perdreau-Remington
F
Götz
F
Characterization of Tn917 insertion mutants of Staphylococcus epidermidis affected in biofilm formation
Infect Immun
 , vol. 
64
 (pg. 
227
-
82
)
8
Heilmann
C
Hussain
M
Peters
G
Götz
F
Evidence for autolysin-mediated primary attachment of Staphylococcus epidermidis to a polystyrene surface
Mol Microbiol
 , vol. 
24
 (pg. 
1013
-
24
)
9
Mack
D
Fischer
W
Krokotsch
A
, et al.  . 
The intercellular adhesin involved in biofilm accumulation of Staphylococcus epidermidis is a linear β-1, 6-linked glucosaminoglycan: purification and structural analysis
J Bacteriol
 , vol. 
178
 (pg. 
175
-
83
)
10
Mack
D
Nedelmann
M
Krokotsch
A
Schwarzkopf
A
Heesemann
J
Characterization of transposon mutants of biofilm-producing Staphylococcus epidermidis impaired in the accumulative phase of biofilm production: genetic identification of a hexosamine-containing polysaccharide intercellular adhesin
Infect Immun
 , vol. 
62
 (pg. 
3244
-
53
)
11
Mack
D
Siemssen
N
Laufs
R
Parallel induction by glucose of adherence and a polysaccharide antigen specific for plastic-adherent Staphylococcus epidermidis: evidence for functional relation to intercellular adhesion
Infect Immun
 , vol. 
60
 (pg. 
2048
-
57
)
12
Fey
PD
Ulphani
JS
Götz
F
Heilmann
C
Mack
D
Characterization of the relationship between polysaccharide intercellular adhesin and hemagglutination in Staphylococcus epidermidis
J Infect Dis
 , vol. 
179
 (pg. 
1561
-
4
)
13
Rupp
ME
Ulphani
JS
Fey
PD
Bartscht
K
Mack
D
Characterization of the importance of polysaccharide intercellular adhesin/hemagglutinin of Staphylococcus epidermidis in the pathogenesis of biomaterial-based infection in a mouse foreign body infection model
Infect Immun
 , vol. 
67
 (pg. 
2627
-
32
)
14
Rupp
ME
Ulphani
JS
Fey
PD
Mack
D
Characterization of Staphylococcus epidermidis polysaccharide intercellular adhesion/hemagglutinin in the pathogenesis of intravascular catheter-associated infection in a rat model
Infect Immun
 , vol. 
67
 (pg. 
2656
-
9
)
15
Heilmann
C
Schweitzer
O
Gerke
C
Vanittanakom
N
Mack
D
Molecular basis of intercellular adhesion in the biofilm-forming Staphylococcus epidermidis
Mol Microbiol
 , vol. 
20
 (pg. 
1083
-
91
)
16
Ulphani
JS
Rupp
ME
Model of Staphylococcus aureus central venous catheter–associated infection in rats
Lab Anim Sci
 , vol. 
49
 (pg. 
283
-
7
)
17
Christensen
GD
Baldassarri
L
Simpson
WA
Bisno
AL
Waldvogel
FA
Colonization of medical devices by coagulase-negative staphylococci
Infections associated with indwelling medical devices
 , 
1994
2d ed
Washington, DC
American Society for Microbiology
(pg. 
45
-
78
)
18
Gristina
AG
Biomaterial-centered infection: microbial adhesion versus tissue integration
Science
 , vol. 
237
 (pg. 
1588
-
95
)
19
Peters
G
Locci
R
Pulverer
G
Adherence and growth of coagulase-negative staphylococci on surfaces of intravenous catheters
J Infect Dis
 , vol. 
146
 (pg. 
479
-
82
)
20
Tojo
M
Yamashita
N
Goldmann
DA
Pier
GB
Isolation and characterization of a capsular polysaccharide adhesin from Staphylococcus epidermidis
J Infect Dis
 , vol. 
157
 (pg. 
713
-
22
)
21
Kojima
Y
Tojo
M
Goldmann
DA
Tosteson
TD
Pier
GB
Antibody to the capsular polysaccharide/adhesin protects rabbits against catheter-related bacteremia due to coagulase-negative staphylococci
J Infect Dis
 , vol. 
162
 (pg. 
435
-
41
)
22
Shiro
H
Muller
E
Gutierrez
N
Transposon mutants of Staphylococcus epidermidis deficient in elaboration of capsular polysaccharide/adhesin and slime are avirulent in a rabbit model of endocarditis
J Infect Dis
 , vol. 
169
 (pg. 
1042
-
9
)
23
McKenney
D
Hubner
J
Muller
E
Wang
Y
Goldmann
DA
The ica locus of Staphylococcus epidermidis encodes production of the capsular polysaccharide/adhesin
Infect Immun
 , vol. 
66
 (pg. 
4711
-
20
)
24
Timmerman
CP
Fleer
A
Besnier
JM
De Graff
L
Cremers
F
Characterization of a proteinaceous adhesin of Staphylococcus epidermidis which mediates attachment to polystyrene
Infect Immun
 , vol. 
59
 (pg. 
4187
-
92
)
25
Veenstra
GJ
Cremers
FM
Van Dijk
H
Fleer
A
Ultrastructural organization and regulation of a biomaterial adhesin of Staphylococcus epidermidis
J Bacteriol
 , vol. 
178
 (pg. 
537
-
41
)
26
Hussain
M
Herrmann
M
Von Eiff
C
Perdreau-Remington
F
Peters
G
A 140-kilodalton extracellular protein is essential for the accumulation of Staphylococcus epidermidis strains on surfaces
Infect Immun
 , vol. 
65
 (pg. 
519
-
24
)
27
Perdreau-Remington
F
Sande
MA
Peters
G
Chambers
HF
The abilities of a Staphylococcus epidermidis wild-type strain and its slime-negative mutant to induce endocarditis in rabbits are comparable
Infect Immun
 , vol. 
66
 (pg. 
2778
-
81
)
Presented in part: 38th Interscience Conference on Antimicrobial Agents and Chemotherapy, San Diego, September 1998 (abstract B-23)
This work was conducted in compliance with institutional guidelines regarding animal studies
Financial support: American Heart Association (grant-in-aid 96006810 to M.E.R.)